Decision Rules on Damaged House Data for Natural Disasters Using The If-Then Method Of The Rough Set Theory

Abstract

Natural disasters remain a real threat for Indonesia. One province in Indonesia which has a high disaster risk is Nusa Tenggara Barat. The impact of disasters can be understood better if the data shape systematically. Data mining techniques that are used in this research is the Rough Set theory. It is an alternative analysis method for analyzing the imperfect data. Disaster that cause damaged houses in Nusa Tenggara Barat, including inconsistent data (rough). Researchers want to know the rules of data patterns (decision rules) using if-then rules from rough set theory. Based on the analysis, gained some rules of data patterns cause heavy damaged houses, minor and minor and heavy damaged houses. 32.75% of the houses were heavily damaged happens when a tornado disaster-stricken area with rain intensity is very low, flat topography and medium density of population. 89.93% of homes with minor damage occurs when an area flooded with low rainfall intensity, undulating topography of the area and the condition of the population is very dense. 38.45% of homes with minor and heavy damage occurred when a severe flood-stricken area with low rainfall intensity, topography of the area is very steep and very dense population conditions.